INDEX

Explanations

brackets

np_max-act · gemini-2.0-flash

The neuron flags ordinary running‐text words in free‐form narrative (dialogue, transcript, or scientific abstract) and stays silent on purely structured metadata or bullet lists.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Swinger

-0.07

 Deadpool

-0.06

 Character

-0.06

 Друг

-0.06

個

-0.06

 سخ

-0.06

fot

-0.06

िक

-0.06

싱

-0.06

 Además

-0.06

POSITIVE LOGITS

-equiv

0.07

两

0.07

 dejtings

0.07

’ya

0.07

ора

0.07

“(

0.07

	Check

0.07

 SETUP

0.07

_sprites

0.06

(sa

0.06

Activations Density 0.016%

brackets

The neuron flags ordinary running‐text words in free‐form narrative (dialogue, transcript, or scientific abstract) and stays silent on purely structured metadata or bullet lists.

No Comments

No Known Activations